A Comparative Study of Time Frequency Representation Techniques for Freeze of Gait Detection and Prediction

• Published in: Sensors (Basel, Switzerland) Vol. 21; no. 19; p. 6446
• Main Authors: Ashfaque Mostafa, Tahjid, Soltaninejad, Sara, McIsaac, Tara L, Cheng, Irene
• Format: Journal Article
• Language: English
• Published: Switzerland MDPI AG 27.09.2021; MDPI
• Subjects: Accelerometers; Activities of daily living; Artificial intelligence; Collaboration; Comparative studies; Computer architecture; Deep learning; detection and predication; ensemble learning; freeze of gait; Gait; Gait Disorders, Neurologic - diagnosis; Gait recognition; Humans; Monitoring systems; Movement disorders; Network latency; Neural networks; Parkinson Disease - diagnosis; Parkinson's disease; Patients; Quality of Life; Representations; Sensors; Time-frequency analysis; Wavelet Analysis; Wavelet transforms; wearable sensor data; detection and predication; ensemble learning; deep learning; wearable sensor data; Parkinson’s disease; freeze of gait
• ISSN: 1424-8220; 1424-8220
• DOI: 10.3390/s21196446

Freezing of Gait (FOG) is an impairment that affects the majority of patients in the advanced stages of Parkinson's Disease (PD). FOG can lead to sudden falls and injuries, negatively impacting the quality of life for the patients and their families. Rhythmic Auditory Stimulation (RAS) can be used to help patients recover from FOG and resume normal gait. RAS might be ineffective due to the latency between the start of a FOG event, its detection and initialization of RAS. We propose a system capable of both FOG prediction and detection using signals from tri-axial accelerometer sensors that will be useful in initializing RAS with minimal latency. We compared the performance of several time frequency analysis techniques, including moving windows extracted from the signals, handcrafted features, Recurrence Plots (RP), Short Time Fourier Transform (STFT), Discreet Wavelet Transform (DWT) and Pseudo Wigner Ville Distribution (PWVD) with Deep Learning (DL) based Long Short Term Memory (LSTM) and Convolutional Neural Networks (CNN). We also propose three Ensemble Network Architectures that combine all the time frequency representations and DL architectures. Experimental results show that our ensemble architectures significantly improve the performance compared with existing techniques. We also present the results of applying our method trained on a publicly available dataset to data collected from patients using wearable sensors in collaboration with A.T. Still University.